Abstract
Objective:In order to realize the intelligent grasping of tobacco stem in tobacco grading process, prevent the manipulator in the intelligent tobacco grading system from damaging the leaf surface during grasping tobacco leaves, and reduce the manual operation in the production of intelligent tobacco grading equipment.Methods:An automatic tobacco stem identification and location model based on improved YOLOv3 convolution neural network was proposed for the identification and classification of single tobacco leaf and the storage of corresponding single tobacco leaf in tobacco grading system. The model changed the structure of the unit module and introduced the attention mechanism module based on the original YOLOv3 model, which optimized the model parameters and used swish activation function to realize the target location and recognition of all the information of tobacco leaf images, and then the tobacco stem target detection model was constructed.Results:The results showed that the loss of improved YOLOv3 model could converge faster, with its mAP increased from 90.46% to 97.48% and its accuracy increased from 95.33% to 97.35%; its regression rate increased from 84.65% to 95.65%, which laid the foundation for the automatic classification of tobacco leaves.Conclusion:Compared with YOLOv3, Faster-rcnn, YOLOv4, Efficientdet algorithm, the proposed algorithm is lighter and more effective. It can reduce the hardware configuration requirements of tobacco stem test platform, improve the economic benefits of tobacco classification system, and provide accurate location information for tobacco feeding and storehouse separation in tobacco classification system.
Publication Date
7-7-2022
First Page
103
Last Page
109
DOI
10.13652/j.spjx.1003.5788.2022.90022
Recommended Citation
Xin-yu, LIU; Tong-meng, HAO; Hong-tao, ZHANG; and Xin-yan, LU
(2022)
"Cigarette stem identification and location method based on improved YOLOv3 network,"
Food and Machinery: Vol. 38:
Iss.
3, Article 17.
DOI: 10.13652/j.spjx.1003.5788.2022.90022
Available at:
https://www.ifoodmm.cn/journal/vol38/iss3/17
References
[1] 王欣,卢俊,徐智,等.基于BP神经网络算法的烟草机械塑料齿轮早期故障监测与优化[J].塑料科技,2021,49(2):91-94.WANG Xin,LU Jun,XU Zhi,et al.Early fault monitoring and optimization of plastic gears in tobacco machinery based on BP neural network algorithm[J].Plastics Science and Technology,2021,49(2):91-94.
[2] 王戈,丁冉,徐玮杰,等.计算机视觉和智能识别技术在烤烟烟叶分级中的应用[J].计算机与应用化学,2019,36(5):548-553.WANG Ge,DING Ran,XU Wei-jie,et al.Application of computer vision and intelligent recognition technology in flue-cured tobacco classification[J].Computer and Applied Chemistry,2019,36(5):548-553.
[3] 王天旺,解立明,刘文,等.基于改进鲁棒多分类SVM的烟叶颜色分级分类方法研究[J].机电信息,2021(5):55-57.WANG Tian-wang,XIE Li-ming,LIU Wen,et al.Study on tobacco color classification method based on improved robust multi-classification SVM[J].Electromechanical Information,2021(5):55-57.
[4] 李峥,王建峰,程小强,等.基于BP神经网络的烤烟外观质量预测模型[J].西南农业学报,2019,32(3):653-658.LI Zheng,WANG Jian-feng,CHENG Xiao-qiang,et al.Prediction model of flue-cured tobacco appearance quality based on BP neural network[J].Journal of Southwest Agricultural Sciences,2019,32(3):653-658.
[5] 雷建生,李再,冉宝新,等.基于SOM神经网络的制丝生产线设备故障趋势辨识方法[J].烟草科技,2019,52(2):109-114.LEI Jian-sheng,LI Zai,RAN Bao-xin,et al.Fault trend identification method of silk production line equipment based on SOM neural network[J].Tobacco Science and Technology,2019,52(2):109-114.
[6] 邵惠芳,赵昕宇,许自成,等.基于SOFM网络的烤烟感官质量聚类模式分析[J].中国烟草学报,2016,22(1):13-23.SHAO Hui-fang,ZHAO Xin-yu,XU Zi-cheng,et al.Cluster model analysis of flue-cured tobacco sensory quality based on SOFM network[J].Journal of China Tobacco Journal,2016,22(1):13-23.
[7] YU Y,ZHANG K L,YANG L,et al.Fruit detection for strawberry harvesting robot in non-structural environment based on Mask-RCNN[J].Computers and Electronics in Agriculture,2019,163:104846.
[8] 朱文魁,刘斌,毛伟俊,等.基于低能X射线透射成像的打叶片烟中烟梗在线检测[J].烟草科技,2015,48(2):69-74.ZHU Wen-kui,LIU Bin,MAO Wei-jun,et al.On-line detection of tobacco stems in threshing tobacco based on low-energy X-ray transmission imaging[J].Tobacco Science and Technology,2015,48(2):69-74.
[9] 崔云月,管一弘,孙娜,等.BP神经网络在烟梗长短梗率检测中的应用[J].软件导刊,2021,20(2):63-67.CUI Yun-yue,GUAN Yi-hong,SUN Na,et al.Application of BP neural network in detection of long and short stem rate of tobacco stems[J].Software Guide,2021,20(2):63-67.
[10] 宋洋,王新,明军.MPC08SP运动控制卡在烟把智能定位系统中的应用[J].计算机技术与发展,2010,20(12):185-188.SONG Yang,WANG Xin,MING Jun.Application of MPC08SP motion control card in intelligent cigarette handle positioning system[J].Computer Technology and Development,2010,20(12):185-188.
[11] 席建平,易浩,刘斌,等.基于FPGA的烟梗在线检测系统设计[J].中国烟草学报,2016,22(5):50-54.XI Jian-ping,YI Hao,LIU Bin,et al.Design of tobacco stem online detection system based on FPGA[J].Journal of China Tobacco Journal,2016,22(5):50-54.
[12] 汤龙.基于透光性差异的烟梗检测分析及算法实现[J].景德镇学院学报,2017,32(6):26-29.TANG Long.Tobacco stem detection analysis and algorithm realization based on light transmittance difference[J].Journal of Jingdezhen University,2017,32(6):26-29.
[13] 郑茜,夏自龙,袁海霞,等.高频阶梯式烟梗分选筛的设计与应用[J].食品与机械,2019,35(7):124-127.ZHENG Xi,XIA Zi-long,YUAN Hai-xia,et al.Design and application of high frequency stepped tobacco stem sorting screen[J].Food & Machinery,2019,35(7):124-127.
[14] REN S,HE K,GIRSHICK R,et al.Faster R-CNN:Towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(6):1 137-1 149.
[15] 梁煜,李佳豪,张为,等.嵌入中心点预测模块的Yolov3遮挡人员检测网络[J].天津大学学报(自然科学与工程技术版),2021,54(5):517-525.LIANG Yu,LI Jia-hao,ZHANG Wei,et al.Yolov3 occlusion personnel detection network embedded with central point prediction module[J].Journal of Tianjin University(Natural Science and Engineering Technology Edition),2021,54(5):517-525.
[16] 卢官有,顾正弘.改进的YOLOv3安检包裹中危险品检测算法[J].计算机应用与软件,2021,38(1):197-204.LU Guan-you,GU Zheng-hong.Improved YOLOv3 detection algorithm for dangerous goods in security parcels[J].Computer Application and Software,2021,38(1):197-204.
[17] 李澎林,章军伟,李伟.基于光流改进与YOLOv3的烟雾检测方法[J].浙江工业大学学报,2021,49(1):9-15.LI Pen-lin,ZHANG Jun-wei,LI Wei.Smoke detection method based on optical flow improvement and YOLOv3[J].Journal of Zhejiang University of Technology,2021,49(1):9-15.
[18] LIU J,WANG X.Correction to:Early recognition of tomato gray leaf spot disease based on MobileNetv2-YOLOv3 model[J].Plant Methods,2021,17(1):19.
[19] YANG F,YANG D,HE Z,et al.Automobile fine-grained detection algorithm based on multi-improved YOLOv3 in smart streetlights[J].Algorithms,2020,13(5):114.
[20] XUE J,LI Z,FUKUDA M,et al.Garbage detection using YOLOv3 in nakanoshima challenge[J].Journal of Robotics and Mechatronics,2020,32(6):1 200-1 210.